The overall goals of this project are to analyze and compare the mechanisms underlying the two major forms of associative learning, classical and operant conditioning. The proposed studies will focus on feeding behavior, which can be modified by both forms of learning, and which is amenable to cellular analyses. The Specific Aims of the proposed studies include: 1. Investigate the cellular mechanisms of operant conditioning. For mechanistic analyses, a previously developed in vitro analogue of operant conditioning was reduced to a single cell (B51) in culture. This single-cell analogue induced changes in B51 that are similar to those induced by in vivo and in vitro conditioning. Aim 1 will: 1) Characterize the modulation of membrane currents by contingent reinforcement; 2) Investigate the intracellular signals that mediate contingent reinforcement; and 3) Confirm that the cellular mechanisms elucidated in the single-cell analogue also occur in the ganglia and assess the contribution of changes in B51 to changes in the function of the feeding circuitry. 2. Characterize the role of B51 in classical conditioning. B51 is a locus of plasticity common to in vivo and in vitro operant and to in vivo classical conditioning. To examine the mechanisms underlying changes in B51 following classical conditioning, it is necessary to use the in vitro analogue. Thus, Aim 2 will: 1) Complete the analysis of changes in B51 that are induced by in vivo training; 2) Confirm that the in vitro analogue of classical conditioning induces changes in B51 similar to those following in vivo conditioning; and 3) Investigate which second-messenger systems mediate the modulation of B51 and asses the contribution of changes in B51 to changes in the function of the feeding circuitry. 3. Identify and analyze additional sites of plasticity that contribute to classical and/or operant conditioning. Although several sites of plasticity have been identified, other sites are likely to contribute to classical and operant conditioning. Thus, Aim 3 will: 1) Examine whether the in vitro analogues of classical and operant conditioning produce changes in the cellular and synaptic properties of sensory, command, pattern-initiating, and pattern-switching neurons in the feeding circuitry; and 2) Confirm that any changes produced by the in vitro analogues are also produced by in vivo training and examine the extent to which the changes are correlated with the behavioral modifications. [unreadable] [unreadable]